Re: [PATCH bpf-next v1 RESEND 5/5] x86: use register_text_tail_vm

[Song Liu <song@kernel.org>]

On Wed, Nov 2, 2022 at 3:24 PM Edgecombe, Rick P
<rick.p.edgecombe@intel.com> wrote:
>
> On Mon, 2022-10-31 at 15:25 -0700, Song Liu wrote:
> > Allocate 2MB pages up to round_up(_etext, 2MB), and register memory
> > [round_up(_etext, 4kb), round_up(_etext, 2MB)] with
> > register_text_tail_vm
> > so that we can use this part of memory for dynamic kernel text (BPF
> > programs, etc.).
> >
> > Here is an example:
> >
> > [root@eth50-1 ~]# grep _etext /proc/kallsyms
> > ffffffff82202a08 T _etext
> >
> > [root@eth50-1 ~]# grep bpf_prog_ /proc/kallsyms  | tail -n 3
> > ffffffff8220f920 t
> > bpf_prog_cc61a5364ac11d93_handle__sched_wakeup       [bpf]
> > ffffffff8220fa28 t
> > bpf_prog_cc61a5364ac11d93_handle__sched_wakeup_new   [bpf]
> > ffffffff8220fad4 t
> > bpf_prog_3bf73fa16f5e3d92_handle__sched_switch       [bpf]
> >
> > [root@eth50-1 ~]#  grep 0xffffffff82200000
> > /sys/kernel/debug/page_tables/kernel
> > 0xffffffff82200000-
> > 0xffffffff82400000     2M     ro   PSE         x  pmd
> >
> > ffffffff82200000-ffffffff82400000 is a 2MB page, serving kernel text,
> > and
> > bpf programs.
> >
> > Signed-off-by: Song Liu <song@kernel.org>
> > ---
> >  arch/x86/include/asm/pgtable_64_types.h | 1 +
> >  arch/x86/mm/init_64.c                   | 4 +++-
> >  include/linux/vmalloc.h                 | 4 ++++
> >  3 files changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/x86/include/asm/pgtable_64_types.h
> > b/arch/x86/include/asm/pgtable_64_types.h
> > index 04f36063ad54..c0f9cceb109a 100644
> > --- a/arch/x86/include/asm/pgtable_64_types.h
> > +++ b/arch/x86/include/asm/pgtable_64_types.h
> > @@ -101,6 +101,7 @@ extern unsigned int ptrs_per_p4d;
> >  #define PUD_MASK     (~(PUD_SIZE - 1))
> >  #define PGDIR_SIZE   (_AC(1, UL) << PGDIR_SHIFT)
> >  #define PGDIR_MASK   (~(PGDIR_SIZE - 1))
> > +#define PMD_ALIGN(x) (((unsigned long)(x) + (PMD_SIZE - 1)) &
> > PMD_MASK)
> >
> >  /*
> >   * See Documentation/x86/x86_64/mm.rst for a description of the
> > memory map.
> > diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
> > index 3f040c6e5d13..5b42fc0c6099 100644
> > --- a/arch/x86/mm/init_64.c
> > +++ b/arch/x86/mm/init_64.c
> > @@ -1373,7 +1373,7 @@ void mark_rodata_ro(void)
> >       unsigned long start = PFN_ALIGN(_text);
> >       unsigned long rodata_start = PFN_ALIGN(__start_rodata);
> >       unsigned long end = (unsigned long)__end_rodata_hpage_align;
> > -     unsigned long text_end = PFN_ALIGN(_etext);
> > +     unsigned long text_end = PMD_ALIGN(_etext);
> >       unsigned long rodata_end = PFN_ALIGN(__end_rodata);
> >       unsigned long all_end;
>
> Check out is_errata93(). Right now it assumes all text is between text-
> etext and MODULES_VADDR-MODULES_END. It's a quite old errata, but it
> would be nice if we had a is_text_addr() helper or something. To help
> keep track of the places where text might pop up.
>
> Speaking of which, it might be nice to update
> Documentation/x86/x86_64/mm.rst with some hints that this area exists.
>
> >
> > @@ -1414,6 +1414,8 @@ void mark_rodata_ro(void)
> >                               (void *)rodata_end, (void *)_sdata);
> >
> >       debug_checkwx();
> > +     register_text_tail_vm(PFN_ALIGN((unsigned long)_etext),
> > +                           PMD_ALIGN((unsigned long)_etext));
> >  }
> >
> >  int kern_addr_valid(unsigned long addr)
> > diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
> > index 9b2042313c12..7365cf9c4e7f 100644
> > --- a/include/linux/vmalloc.h
> > +++ b/include/linux/vmalloc.h
> > @@ -132,11 +132,15 @@ extern void vm_unmap_aliases(void);
> >  #ifdef CONFIG_MMU
> >  extern void __init vmalloc_init(void);
> >  extern unsigned long vmalloc_nr_pages(void);
> > +void register_text_tail_vm(unsigned long start, unsigned long end);
> >  #else
> >  static inline void vmalloc_init(void)
> >  {
> >  }
> >  static inline unsigned long vmalloc_nr_pages(void) { return 0; }
> > +void register_text_tail_vm(unsigned long start, unsigned long end)
> > +{
> > +}
> >  #endif
>
> This looks like it should be in the previous patch.

Good catch! I will fix it in the next version.

Thanks,
Song

>
> >
> >  extern void *vmalloc(unsigned long size) __alloc_size(1);